324 



Professor J. Tyndall, 



[Jan. 23, 



there ought to be a whole year's unfrozen fissures in the ice. 

 Such fissures surely would not require the act of freezing to render 

 them visible ; they would be seen when filled with blue water just 

 as well as when filled with blue ice. But they never have been 

 observed ; and it is therefore to be inferred that they have no 

 existence. With regard to the drag towards the centre, which is 

 supposed to arise from the viscosity, and in which direction it is 

 stated that " filaments slide past each other," it is by no means 

 clear on mechanical grounds that such a drag exists. For the 

 transfer of matter from the sides to the centre, in consequence of 

 such a drag, must finally absorb the former, unless to make good the 

 loss a motion in some other portion of the glacier from the centre to 

 the sides, that is in a direction opposed to the theory, be established. 

 Let the line AB, (Fig. 1) represent the centre of a glacier; C D 

 its side, and a a point b6tween both. Draw mn, ojp, making 



JH^ 



Fig 



equal angles with the centre and side. In consequence of the 

 quicker flow of the centre the line m n tends to shorten itself, 

 causing a thrust on the point «, which urges it towards the side 

 C D ; the line op for the same reason tends to elongate itself, 

 which produces a drag of the point a towards the centre. The 

 point a is here solicited by two equal forces, and the resultant 

 motion will be along the line A B, parallel to the length of the 

 glacier. This result receives the most complete confirmation from 

 observation, so that the drag towards the centre expresses only half 

 the conditions of the problem. 



To test the question on a small scale, the following experiment 



Fig. 2. 



O 



was made, A B C D, (Fig. 2) is a wooden trough, six feet long, and 

 one foot wide. The end, A C, of the trough is elevated. O is a 



